User Profile Linking

1. A method, comprising: receiving, from a first device, audio data corresponding to a spoken natural language input; performing speech processing on the audio data to generate natural language understanding (NLU) results data representing the spoken natural language input; determining a first component configured to execute with respect to the NLU results data to perform an action responsive to the spoken natural language input; sending, to the first component, the NLU results data; receiving, from the first component, first data requesting a user profile token corresponding to a user profile stored by the first component; determining a first user profile corresponding to the audio data, the first user profile being stored by a natural language processing (NLP) system; determining a user-specific information type required by the first component to generate a user profile storable by the first component; querying the NLP profile storage for user-specific information, corresponding to the user-specific information type, represented in the first user profile; receiving, from the NLP profile storage, first user-specific information corresponding to the user-specific information type; sending, to the first component, the first user-specific information; receiving, from the first component, a first user profile token corresponding to a second user profile generated by the first component using the first user-specific information; and storing, in the first user profile, an association between the first component and the first user profile token.

2. The method of claim 1 , further comprising: determining first data, received from a developer device, representing an intent with respect to which the first component requires a first component-stored user profile to execute; and determining the intent is represented in the NLU results data; and wherein the user-specific information type is determined after determining the intent is represented in the first data and the NLU results data.

3. The method of claim 1 , further comprising: receiving, from the first device, second audio data corresponding to a second spoken natural language input; performing speech processing on the second audio data to generate second NLU results data representing the second spoken natural language input; determining a second component configured to execute with respect to the second NLU results data to perform a second action responsive to the second spoken natural language input; sending, to the second component, the second NLU results data; receiving, from the second component, second data requesting a user profile token corresponding to a user profile stored by the second component; determining the first user profile corresponds to the second audio data; after determining the first user profile corresponds to the second audio data, determining a second user-specific information type required by the second component to generate a user profile storable by the first component; after determining the first user profile corresponds to the second audio data, determining a third user-specific information type required by the second component to generate a user profile storable by the second component; querying the NLP profile storage for user-specific information, corresponding to the second user-specific information type and the third user-specific information type, represented in the first user profile; receiving, from the NLP profile storage, second user-specific information corresponding to the second user-specific information type; sending, to the first device, third data requesting user-specific information corresponding to the third user-specific information type; receiving, from the first device, third user-specific information corresponding to the third user-specific information type; sending, to the second component, the second user-specific information and the third user-specific information; receiving, from the second component, a second user profile token corresponding to a third user profile generated by the second component using the second user-specific information and the third user-specific information; and storing, in the first user profile, a second association between the second component and the second user profile token.

4. The method of claim 1 , further comprising: receiving, from the first device, an indicator representing a touch input was detected while a first application was executed, the touch input representing a user profile token is to be sent to the first device, the user profile token corresponding to a user profile stored by the NLP system for the first component; and sending, to the first device, the user profile token, wherein the sending of the user profile token causes the first application to send the user profile token to a second application installed on the first device, the second application being associated with the first component, wherein receipt of the user profile token by the first component results in the first component generating a first component-stored user profile corresponding to the user profile token.

5. A natural language processing (NLP) system, comprising: at least one processor; and at least one memory comprising instructions that, when executed by the at least one processor, cause the NLP system to: determine natural language understanding (NLU) results data representing a natural language input; determine a first component configured to execute with respect to the NLU results data to perform an action responsive to the natural language input; send, to the first component, the NLU results data; receive, from the first component, first data requesting a user profile token corresponding to a user profile stored by the first component; determine a user-specific information type required by the first component to generate a user profile storable by the first component; determine a first user profile corresponding to the natural language input, the first user profile being stored by the NLP system; determine, in the first user profile, first user-specific information corresponding to the user-specific information type; and send, to the first component, the first user-specific information.

6. The NLP system of claim 5 , wherein the at least one memory further comprises instructions that, when executed by the at least one processor, further cause the NLP system to: query a NLP profile storage for a user profile token, corresponding to a user profile stored by the first component, represented in the first user profile; and receive, from the NLP profile storage, a no results query response, wherein the user-specific information type is determined after receiving the no results query response.

7. The NLP system of claim 5 , wherein the at least one memory further comprises instructions that, when executed by the at least one processor, further cause the NLP system to: after sending the first user-specific information to the first component, receive, from the first component, a first user profile token corresponding to a second user profile generated by the first component using the first user-specific information; and store, in the first user profile, an association between the first component and the first user profile token.

8. The NLP system of claim 5 , wherein the at least one memory further comprises instructions that, when executed by the at least one processor, further cause the NLP system to: determine second data, received from a developer device, representing an intent with respect to which the first component requires a first component-stored user profile to execute; and determine the intent is represented in the NLU results data, wherein the user-specific information type is determined based at least in part on determining the intent is represented in the second data and the NLU results data.

9. The NLP system of claim 5 , wherein the at least one memory further comprises instructions that, when executed by the at least one processor, further cause the NLP system to: determine a second user-specific information type required by the first component to generate a user profile storable by the first component; determine user-specific information, corresponding to the second user-specific information type, is missing from the first user profile; send, to a first device associated with the first user profile, second data requesting user-specific information corresponding to the second user-specific information type; receive, from the first device, second user-specific information corresponding to the second user-specific information type; and send, to the first component, the second user-specific information.

10. The NLP system of claim 5 , wherein the at least one memory further comprises instructions that, when executed by the at least one processor, further cause the NLP system to: determine second NLU results data representing a second natural language input; determine an intent, represented in the second NLU results data, represents a second component is to be enabled to execute with respect to future natural language inputs corresponding to the first user profile; determine a second user-specific information type required by the second component to generate a user profile storable by the second component; determine, in the first user profile, second user-specific information corresponding to the second user-specific information type; and send, to the second component, the second user-specific information.

11. The NLP system of claim 5 , wherein the at least one memory further comprises instructions that, when executed by the at least one processor, further cause the NLP system to: receive, from the first component, second data representing the first component requires the user-specific information type to generate a user profile storable by the first component; determine third data received from a skill developer device prior to the first component being approved for runtime processing of natural language inputs; and determine the third data includes the user-specific information type, wherein the first user-specific information is determined after determining the third data includes the user-specific information type.

12. The NLP system of claim 5 , wherein the at least one memory further comprises instructions that, when executed by the at least one processor, further cause the NLP system to: receive, from a first device, an indicator representing a touch input was detected while a first application was executed, the touch input representing a second user profile token is to be sent to the first device, the user profile token corresponding to a user profile stored by the NLP system for the first component; and send, to the first device, the second user profile token, wherein the sending of the second user profile token causes the first application to send the second user profile token to a second application installed on the first device, the second application being associated with the first component, wherein receipt of the second user profile token by the first component results in the first component generating a first component-stored user profile corresponding to the second user profile token.

13. A method, comprising: determining natural language understanding (NLU) results data representing a natural language input; determining a first component configured to execute with respect to the NLU results data to perform an action responsive to the natural language input; sending, to the first component, the NLU results data; receiving, from the first component, first data requesting a user profile token corresponding to a user profile stored by the first component; determining a user-specific information type required by the first component to generate a user profile storable by the first component; determining a first user profile corresponding to the natural language input, the first user profile being stored by a natural language processing (NLP) system; determining, in the first user profile, first user-specific information corresponding to the user-specific information type; and sending, to the first component, the first user-specific information.

14. The method of claim 13 , further comprising: querying a NLP profile storage for a user profile token, corresponding to a user profile stored by the first component, represented in the first user profile; and receiving, from the NLP profile storage, a no results query response, wherein the user-specific information type is determined after receiving the no results query response.

15. The method of claim 13 , further comprising: after sending the first user-specific information to the first component, receiving, from the first component, a first user profile token corresponding to a second user profile generated by the first component using the first user-specific information; and storing, in the first user profile, an association between the first component and the first user profile token.

16. The method of claim 13 , further comprising: determining second data, received from a developer device, representing an intent with respect to which the first component requires a first component-stored user profile in order to execute; and determining the intent is represented in the NLU results data, wherein the user-specific information type is determined based at least in part on determining the intent is represented in the second data and the NLU results data.

17. The method of claim 13 , further comprising: determining a second user-specific information type required by the first component to generate a user profile storable by the first component; determining user-specific information, corresponding to the second user-specific information type, is missing from the first user profile; sending, to a first device associated with the first user profile, second data requesting user-specific information corresponding to the second user-specific information type; receiving, from the first device, second user-specific information corresponding to the second user-specific information type; and sending, to the first component, the second user-specific information.

18. The method of claim 13 , further comprising: determining second NLU results data representing a second natural language input; determining an intent, represented in the second NLU results data, represents a second component is to be enabled to execute with respect to future natural language inputs corresponding to the first user profile; determining a second user-specific information type required by the second component to generate a user profile storable by the second component; determining, in the first user profile, second user-specific information corresponding to the second user-specific information type; and sending, to the second component, the second user-specific information.

19. The method of claim 13 , further comprising: receiving, from the first component, second data representing the first component requires the user-specific information type to generate a user profile storable by the first component; determining third data received from a skill developer device prior to the first component being approved for runtime processing of natural language inputs; and determining the third data includes the user-specific information type, wherein the first user-specific information is determined after determining the third data includes the user-specific information type.

20. The method of claim 13 , further comprising: receiving, from a first device, an indicator representing a touch input was detected while a first application was executed, the touch input representing a second user profile token is to be sent to the first device, the user profile token corresponding to a user profile stored by the NLP system for the first component; and sending, to the first device, the second user profile token, wherein the sending of the second user profile token causes the first application to send the second user profile token to a second application installed on the first device, the second application being associated with the first component, wherein receipt of the second user profile token by the first component results in the first component generating a first component-stored user profile corresponding to the second user profile token.